Skin Segmentation using Color Pixel Classification: Analysis and Comparison
Document Type
Journal Article
Publisher
IEEE Computer Society
Faculty
Faculty of Computing, Health and Science
School
School of Computer and Information Science / Centre for Communications Engineering Research
RAS ID
3482
Abstract
This work presents a study of three important issues of the color pixel classification approach to skin segmentation: color representation, color quantization, and classification algorithm. Our analysis of several representative color spaces using the Bayesian classifier with the histogram technique shows that skin segmentation based on color pixel classification is largely unaffected by the choice of the color space. However, segmentation performance degrades when only chrominance channels are used in classification. Furthermore, we find that color quantization can be as low as 64 bins per channel, although higher histogram sizes give better segmentation performance. The Bayesian classifier with the histogram technique and the multilayer perceptron classifier are found to perform better compared to other tested classifiers, including three piecewise linear classifiers, three unimodal Gaussian classifiers, and a Gaussian mixture classifier.
DOI
10.1109/TPAMI.2005.17
Access Rights
free_to_read
Comments
Phung, S. L., Bouzerdoum, A., & Chai, D. K. (2005). Skin segmentation using color pixel classification: analysis and comparison. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(1), 148-154. Available here